Introduction

Research background and motivations

The rapid advancement of artificial intelligence (AI) technology has become a major catalyst for global technological progress. AI’s capability to analyze vast amounts of data allows for more accurate market forecasts and risk assessments, thereby aiding management in making better-informed strategic decisions1,2,3. In the realm of Environmental, Social, and Governance (ESG), AI assists enterprises with environmental monitoring, fulfilling social responsibilities, and optimizing governance structures. By employing intelligent methods, AI enhances a company’s ESG performance and promotes sustainable development4,5,6. Despite the growing integration of AI in ESG practices, its effectiveness and specific impacts on the sustainable development of central state-owned enterprises remain underexplored. In China, where central state-owned enterprises are fundamental to the national economy, these enterprises carry substantial responsibilities related to resource utilization, environmental protection, social responsibility, and governance structures. Therefore, investigating the influence of AI-driven ESG practices on the sustainable development performance of these enterprises presents significant academic value and offers practical guidance for implementation.

The significance of ESG metrics for China’s central state-owned enterprises is profound. These metrics are crucial not only for the sustainable development of the enterprises themselves but also for the broader economic and social development of the nation. As pillars of the national economy, central state-owned enterprises bear significant responsibility in resource utilization, environmental protection, social responsibility, and governance structures. ESG metrics not only help central state-owned enterprises achieve economic benefits but also enhance their social and environmental performance, boosting long-term competitiveness and contributing to the country’s high-quality development and comprehensive social progress7,8,9. The application of AI in enhancing ESG performance and promoting corporate sustainable development is showing emerging trends10,11,12. AI plays a vital role in environmental protection. For example, using sensors and data analysis technologies, AI can monitor air quality, water resources, and carbon emissions in real-time. This capability helps enterprises promptly identify and address environmental issues, thereby achieving more refined and scientific environmental management13. In terms of social responsibility, AI can evaluate an enterprise’s performance in labor conditions, supply chain management, and social contributions through big data analysis14. Regarding corporate governance, AI technologies are applied in risk management, compliance monitoring, and decision support systems, enhancing the transparency and efficiency of corporate governance15,16,17. For instance, through natural language processing, AI can automatically review and analyze large volumes of legal and compliance documents, identify potential risks, and provide recommendations, ensuring that enterprises adhere to relevant regulations and standards18. Overall, these applications of AI not only improve an enterprise’s ESG performance but also drive companies toward more sustainable and responsible operations, demonstrating the substantial potential of technological innovation in corporate management19,20,21.

The motivation for studying the relationship between AI-driven ESG practices and the sustainable development of central state-owned enterprises lies in the transformative potential of AI technology. AI excels in data processing, intelligent decision-making, and process optimization, making it a crucial tool for enhancing ESG performance. Therefore, exploring how AI drives ESG practices in central state-owned enterprises is of significant practical and academic value. The innovations of this paper include systematically analyzing the specific applications of AI technology across various ESG domains, thereby addressing the current research gap concerning the comprehensive impact of AI on environmental, social, and governance aspects. Additionally, this study constructs a model to evaluate the impact of AI-driven ESG practices on the sustainable development of central state-owned enterprises, providing detailed data support and a theoretical basis through quantitative analysis and empirical research.

Research objectives

This study aims to comprehensively investigate the application effects of AI in the ESG strategies of central state-owned enterprises and to assess its impact on the sustainable development performance of these enterprises. Specific objectives include analyzing the application and effects of AI in the ESG practices of central state-owned enterprises, exploring its impact on environmental protection, employee welfare, and corporate governance, and quantitatively evaluating the effectiveness of AI in the ESG strategies of central state-owned enterprises through empirical research. Additionally, the direct impact of AI technology on corporate sustainable development performance is evaluated, highlighting its potential to enhance resource utilization efficiency, improve social image, and foster long-term sustainability. Furthermore, the mediating role of ESG performance between AI application and corporate sustainable development performance is examined, elucidating how AI technology can drive corporate sustainability by optimizing ESG practices. By achieving these objectives, this study seeks to provide both theoretical foundations and practical guidance for central state-owned enterprises in developing and implementing AI-driven ESG strategies. This approach aims to enhance environmental performance, social value, and governance efficiency, ultimately laying the groundwork for the attainment of long-term sustainability goals.

Literature review

Overview of ESG performance and corporate sustainable development

ESG performance indicators are pivotal for corporate sustainable development. Environmental performance encompasses various factors, including a company’s environmental footprint, resource utilization efficiency, and carbon emissions management. Social performance addresses the interactions between a company and its stakeholders—such as employees, customers, and suppliers—focusing on aspects like employee welfare, community engagement, and human rights protection. Governance performance emphasizes the robustness and transparency of a company’s internal management mechanisms, including internal audit procedures, risk management protocols, and board oversight22,23,24,25. Bekaert26 noted that firms with high ESG ratings exhibit more stable stock performance and superior long-term investment returns. Moreover, Doni27 found correlations between ESG performance and corporate innovation, employee satisfaction, and brand reputation, positively impacting financial and market performance. Thus, ESG indicators are not only key metrics for measuring sustainable development but also vital factors for companies aiming to achieve long-term competitive advantages and social recognition.

Overview of the application of AI in ESG practices

The application of AI in ESG practices holds significant promise28,29,30. The theory of AI encompasses fields such as machine learning, data mining, and natural language processing, focusing primarily on how computer systems can emulate human intelligence, learn, and adapt to their environments31,32,33. This theoretical framework elucidates the methods and impacts of applying AI technologies within the ESG domain. For instance, machine learning techniques can assist companies in identifying potential environmental risks and social issues within large datasets, providing intelligent solutions. Natural language processing technologies facilitate real-time monitoring and analysis of employee feedback and societal sentiments, enabling companies to promptly adjust their corporate social responsibility strategies34,35,36. Regarding the environment, Aljohani37 observed that AI technology aids companies in real-time environmental data monitoring, risk prediction, and intelligent environmental management through big data analysis and machine learning. Concerning society, Richey38 noted that AI optimizes employee management, enhances satisfaction, and monitors supply chain compliance, thereby improving corporate social responsibility fulfillment. In terms of governance, Bhima39 found that AI enhances internal audit efficiency, risk management accuracy, and decision-making support, bolstering corporate governance structure and transparency. These research findings underscore the pivotal role of AI in ESG practices, providing companies with innovative solutions to enhance their sustainable development performance and social impact.

Overview of the analysis of AI-enhanced ESG practices from global and Chinese perspectives

A comparative analysis of the application of AI to enhance ESG initiatives from both global and Chinese perspectives elucidates both commonalities and disparities, highlighting the potential roles and impact mechanisms of AI within the ESG domain40,41,42. AI technologies can assist companies in identifying and addressing environmental and social issues through data analysis and predictive modeling, thereby enhancing overall ESG performance. Furthermore, the integration of ESG with AI theory underscores the application of AI technologies in corporate governance and risk management, promoting the intelligence and efficiency of internal management practices. Globally, the integration of AI into ESG practices has garnered widespread attention and achieved significant progress. For instance, Onyeaka43 highlighted that companies worldwide commonly employ AI technology to optimize environmental management, augment social responsibility fulfillment, and fortify corporate governance. Similarly, in China, the utilization of AI in the ESG domain is progressively gaining traction. Hu44 observed Chinese enterprises leveraging AI technology to facilitate real-time and intelligent environmental monitoring, improve employee welfare and social engagement, and enhance internal audit and risk management capabilities. Despite evident methodological differences between global and Chinese approaches to utilizing AI for enhancing ESG, both contexts demonstrate a shared commitment to sustainable development and an ongoing trend of exploration.

Research gaps and hypotheses of the study

The current research landscape highlights two primary gaps: Firstly, research on the relationship between AI technology and corporate ESG performance remains underdeveloped. Empirical studies focusing on the specific application mechanisms of AI within the ESG framework of central state-owned enterprises, as well as its effects, are notably lacking. In particular, there is a deficiency of systematic research exploring how AI technology can enhance corporate ESG performance, thereby advancing corporate sustainable development across the three dimensions of environmental protection, social responsibility, and corporate governance optimization. Secondly, the mediating role of ESG performance between AI technology and corporate sustainable development has not been adequately addressed. While existing literature suggests that improvements in ESG performance can significantly drive long-term corporate sustainability, its role in bridging the application of AI technology and sustainable development outcomes has not been thoroughly validated. Few studies examine whether AI technology influences corporate economic performance, social image, and environmental responsibility by improving ESG performance. This theoretical gap restricts the practical implications of AI technology in corporate strategic management and policy formulation.

To address the aforementioned research gaps, this study seeks to systematically evaluate the impact mechanisms of AI technology on the ESG performance and sustainable development of central state-owned enterprises, while further exploring the mediating effect of ESG performance in their relationship. The hypotheses of this study are grounded in the Resource-Based View (RBV) theory and Sustainable Development Theory. The RBV theory posits that a company’s core competitiveness arises from its unique resources and capabilities. As an innovative resource, AI technology can help enterprises enhance management effectiveness and operational efficiency, thereby improving their environmental, social, and governance performance. The Sustainable Development Theory emphasizes that a company’s long-term success hinges not only on financial performance but also on its commitment to social responsibility and environmental protection.

(1) According to the RBV theory, AI technology can assist enterprises in optimizing resource allocation and improving operational efficiency, particularly in the areas of environmental management, social responsibility, and corporate governance. For instance, AI can enhance a company’s effectiveness in resource utilization, emissions control, and other environmental areas through intelligent management systems, thereby improving environmental performance. The application of AI in employee management, health monitoring, and workplace safety can also strengthen the company’s performance in social responsibility. Moreover, AI’s role in risk management and decision support contributes to optimizing the company’s governance structure.

Based on the aforementioned rationale, Hypothesis 1 is proposed: The application of AI technology significantly enhances corporate performance across the three dimensions of ESG.

(2) AI technology, by driving transformation at various levels within enterprises, can not only improve ESG performance but also potentially contribute to an increase in corporate sustainable development performance. Sustainable development performance encompasses multiple aspects, including economic benefits, social responsibility, and environmental impact. By improving operational efficiency and optimizing resource allocation, AI technology may have a long-term effect on environmental, social, and governance outcomes.

Therefore, Hypothesis 2 is proposed: Under other conditions remaining constant, the application of AI technology can effectively enhance the overall sustainable development performance of central state-owned enterprises.

(3) The relationships in Hypotheses 1 and 2 may not be solely direct and linear; corporate ESG performance may play a mediating role between the application of AI technology and corporate sustainable development performance. According to mediation effect theory, the application of AI technology, by enhancing a company’s performance in environmental, social, and governance domains, may indirectly influence its sustainable development outcomes. For example, strong ESG performance not only improves the company’s social image but may also enhance resource allocation efficiency, creating additional market opportunities and competitive advantages.

Therefore, Hypothesis 3 is proposed: ESG plays a mediating role between the application of AI technology and corporate sustainable development performance.

By testing these hypotheses, this study aims to provide both theoretical foundations and practical guidance for central state-owned enterprises in formulating and implementing AI-driven ESG strategies, thereby contributing to the achievement of the company’s long-term sustainable development goals.

Research methodology

This study aims to comprehensively examine the application effects of AI technology in central state-owned enterprises, with particular focus on its impact on the three dimensions of ESG performance, as well as the mediating role of ESG between AI technology and corporate sustainable development performance. The primary objective is to uncover the mechanisms through which AI technology enhances ESG performance, strengthens market competitiveness, and promotes sustainable development within these enterprises. The study quantitatively assesses the influence of AI on the ESG performance of these enterprises to elucidate its role in enhancing market competitiveness and achieving sustainable development. Using data from publicly listed central state-owned enterprises between 2016 and 2022, regression models and mediation effect models are employed to examine the specific impact of AI technology on ESG performance and sustainable development outcomes. By employing a mixed-methods approach that combines qualitative and quantitative analyses, the study will examine how AI technology improves ESG performance and market competitiveness, thereby providing empirical support for sustainable development28,45,46. Through an in-depth analysis of these issues, this study seeks to offer both theoretical and practical guidance for central state-owned enterprises in implementing effective ESG strategies.

Qualitative research design

The qualitative research method facilitates an in-depth understanding of the application of AI technology within central state-owned enterprises. A case study approach is employed, involving the careful selection of several representative central state-owned enterprises as subjects of the study. These enterprises occupy significant positions across various industries, including energy, manufacturing, finance, and technology, ensuring the universality and representativeness of the research findings.

In selecting the cases, publicly available ratings and reports on the ESG performance of the enterprises were considered, alongside their leadership and maturity in AI applications. The objective was to choose enterprises that demonstrate experience and achievements in ESG management and the application of AI technology. Data collection primarily utilized two methods: in-depth interviews and document analysis.

  1. (1)

    In-depth Interviews: Conducting in-depth interviews with the management and key stakeholders of the enterprises to gather insights into their perspectives and experiences regarding the application of AI technology.

  2. (2)

    Document Analysis: Analyzing annual reports, sustainability reports, and official announcements of the enterprises to obtain detailed information about the status of AI project implementation, resource allocation, and application outcomes.

The collected data are coded and classified using content analysis and thematic analysis methods to identify key themes and patterns related to the application of AI technology. This process facilitates an understanding of the effectiveness of AI technology across different ESG domains and how these applications influence the sustainable development performance of enterprises.

Quantitative research design

The quantitative research method facilitates the quantification of the impact of AI technology on the environmental, social, and governance (ESG) performance of central state-owned enterprises. Data are collected from relevant departments and management personnel within these enterprises through a structured questionnaire survey.

The questionnaire is designed to comprehensively address the three dimensions of ESG: corporate governance, environmental protection, and social responsibility. Drawing on previous related research, each dimension is comprised of five questions, resulting in a total of 15 questions. The results are presented in Table 147,48,49.

Table 1 Questionnaire item overview.
Full size table

As shown in Table 1, the questions are designed to assess respondents’ perceptions of the effectiveness of AI technology in ESG practices. For instance, within the corporate governance category, inquiries focused on respondents’ evaluations of the transparency of decision-making processes and the extent to which the company utilizes AI technology for risk management. In the environmental protection category, questions explored how the company employs AI to monitor and manage its environmental impact, as well as the effectiveness of AI in enhancing overall environmental performance. The social responsibility category concentrated on how AI technology improves employee welfare and how it is leveraged to enhance the impact of the company’s social responsibility initiatives.

A questionnaire survey was administered to a sample of enterprises, with 213 questionnaires distributed and 200 valid responses collected, resulting in a response rate of 93.9%. Descriptive statistical analysis of the data was conducted using SPSS 26.0 software to ensure the representativeness of the sample. In addition to demographic information such as gender and age, data were collected on respondents’ professional backgrounds, including their specific roles within central state-owned enterprises. This information aids in understanding the varying perspectives and needs regarding AI technology applications across different departments and highlights any differences in responses among these departments, thereby enhancing the interpretation of the research findings and identifying areas where the benefits of AI are most pronounced.

Descriptive statistical analysis was performed on the questionnaire scores, calculating mean scores and standard deviations for each dimension to assess the performance of central state-owned enterprises across various ESG dimensions.

Construction of the regression model

To systematically investigate the application of AI technology in the ESG strategy of central state-owned enterprises and its impact on corporate sustainable development performance, this study constructs a multiple regression model. The model incorporates the degree of AI technology usage as the explanatory variable, corporate sustainable development performance as the dependent variable, and ESG performance as the mediating variable, while also including a series of control variables to enhance the model’s accuracy.

The explanatory variable, representing the extent of AI technology usage, is measured through text analysis of the frequency of AI-related terms in the annual reports of publicly listed companies. This approach involves constructing an AI lexicon and counting the occurrences of relevant terms within the reports. The natural logarithm of the count, plus one, is used as an indicator to reflect the level of AI application within the enterprise50.

The dependent variable, corporate sustainable development performance, is measured using the Sustainable Development Index (SDI). This index integrates the company’s performance in environmental protection, social responsibility, and economic benefits, providing a comprehensive assessment of the enterprise’s overall sustainable development level.

The mediating variable is corporate ESG performance, measured using the Huazheng ESG Index. This index provides a comprehensive evaluation of a company’s performance across the environmental, social, and governance dimensions, serving as a key reference for assessing ESG levels.

To mitigate omitted variable bias, several control variables are included in the model, such as company size, leverage (Lev) ratio, asset turnover (AT), revenue growth (RG), Tobin’s Q, and firm age. A summary of all variables and their corresponding explanations is presented in Table 2.

Table 2 Variable description and explanation.
Full size table

Based on the analysis, a model is constructed to explore the relationships between AI, ESG, and sustainable development performance in enterprises:

$${ESG}_{i}={\alpha }_{0}+{\alpha }_{1}{AI}_{i}+{\alpha }_{2}{X}_{i}+\sum Year+\sum Industry+\varepsilon $$
(1)
$${SDI}_{i}={\beta }_{0}+{\beta }_{1}{AI}_{i}+{\beta }_{2}{X}_{i}+\sum Year+\sum Industry+\varepsilon $$
(2)
$${SDI}_{i}={\gamma }_{0}+{\gamma }_{1}{AI}_{i}+{\gamma }_{2}{ESG}_{i}+{{\gamma }_{3}X}_{i}+\sum Year+\sum Industry+\varepsilon $$
(3)

Here, \({ESG}_{i}\) represents the ESG performance of the i-th enterprise; \({AI}_{i}\) represents the level of AI technology usage; \({SDI}_{i}\) represents the sustainable development performance of the i-th enterprise; \({X}_{i}\) denotes the set of control variables; \(\varepsilon \) is the random error term. \(\sum Year\) and \(\sum Industry\) represent the control variables for year and industry.

To test the mediating effect of ESG, a mediation model is constructed as shown in Eqs. (4) to (6):

$${ESG}_{i}={\alpha }_{0}+{\alpha }_{1}{AI}_{i}+{X}_{i}+\varepsilon (Path\_a)$$
(4)
$${SDI}_{i}={\beta }_{0}+{\beta }_{1}{AI}_{i}+{X}_{i}+\varepsilon (Path\_b)$$
(5)
$${SDI}_{i}={\gamma }_{0}+{\gamma }_{1}{AI}_{i}+{\gamma }_{2}{ESG}_{i}+{X}_{i}+\varepsilon (Path\_c)$$
(6)

In the model setup of this study, for the mediation effect to be valid, the following conditions must be met: the explanatory variable AI must have a significant effect on the dependent variable SDI; the explanatory variable AI must have a significant effect on the mediating variable ESG; both the explanatory variable AI and the mediating variable ESG must be regressed simultaneously on the dependent variable SDI, and the mediating variable ESG must have a significant effect on the dependent variable SDI.

Experimental design and performance evaluation

Datasets collection

When selecting central state-owned enterprises as research subjects, several criteria must be considered to ensure the representativeness and credibility of the study. The following provides a detailed explanation of the criteria for selecting included in the study:

(1) Industry Representation

Central state-owned enterprises chosen should span diverse industries, including but not limited to energy, manufacturing, finance, and technology. This approach ensures the universality and representativeness of the research results, avoiding confinement to specific circumstances of particular industries.

(2) ESG Performance Level

Priority is given to selecting central state-owned enterprises with higher ESG performance as research subjects. The assessment of ESG performance can be based on publicly available data such as ESG ratings, reports, and sustainable development indices, ensuring that the selected research subjects have a certain level of ESG management and practical experience.

(3) Degree of AI Application

Consider the leadership and maturity of central state-owned enterprises in the application of AI. Select central state-owned enterprises with a certain strength and experience in the application of AI technology as research subjects to ensure comprehensive exploration of the impact of AI on ESG performance.

(4) Data Accessibility

Ensure that the selected central state-owned enterprises have publicly available ESG performance data and relevant corporate information for data collection and analysis. At the same time, central state-owned enterprises are willing to cooperate with the research and provide necessary data support and participation in the study.

To ensure the scientific rigor and consistency of the data, a sample pool was initially constructed by selecting all publicly listed central state-owned enterprises from 2016 to 2022 that met the criteria of publicly disclosed ESG reports and financial data. The sample pool was then stratified by industry category (such as energy, manufacturing, finance, technology, etc.) to ensure a balanced selection of representative enterprises from each industry. Within each industry stratum, the sample was further filtered based on AI usage, leading to the final selection of the study sample. The data for this study spans the period from 2016 to 2022 and includes relevant data from publicly listed central state-owned enterprises, primarily encompassing corporate ESG performance, AI application indicators, and control variable data. The Huazheng ESG Index serves as the primary measure for assessing corporate ESG performance. Compiled by a leading Chinese index provider, this index scores and ranks companies based on their performance across environmental, social, and governance dimensions, offering high authority and reference value. AI usage data is sourced from the annual reports of publicly listed central state-owned enterprises, with natural language processing techniques applied to analyze the text of these reports. The frequency of AI-related terms is calculated to construct a quantitative indicator of AI technology usage. Control variable data is sourced from the Wind Financial Terminal, the China Securities Regulatory Commission (CSRC) database, financial statements disclosed by the Shanghai and Shenzhen stock exchanges, and other relevant sources. Descriptive statistics for all variables are provided in Table 3.

Table 3 Descriptive statistics of sample enterprise data.
Full size table

As presented in Table 3, the average AI usage is 3.54, with a standard deviation of 1.09, indicating some variation in the level of AI technology adoption across different central state-owned enterprises. The maximum value is 5.29, and the minimum value is 1.69, reflecting a wide distribution of AI application among the sample enterprises. The mean of the SDI is 63.02, with a standard deviation of 4.87, suggesting some fluctuation in corporate sustainable development performance. ESG performance, with a mean of 68.26, indicates that the sample enterprises generally perform well in terms of environmental, social, and governance aspects. Additionally, control variables such as company size, leverage ratio, and asset turnover also exhibit variability. Notably, Tobin’s Q has a maximum value of 10.67, implying that some companies in the sample have significantly higher market values and more efficient asset allocations compared to others. Overall, the data distribution is reasonable and provides a robust foundation for subsequent regression analysis and empirical research.

A total of 213 questionnaires were distributed, resulting in the collection of 200 valid responses, yielding a response rate of 93.9%. Statistical analysis was performed utilizing SPSS 26.0 software.

Experimental environment and parameters setting

The study primarily employs support vector machines (SVMs) to analyze and model the collected data, investigating the impact mechanism and effects of AI on the ESG performance of central state-owned enterprises.

The specific steps of SVM modeling are outlined as follows:

(1) Data Preparation

ESG performance data and AI application indicator data are gathered from central state-owned enterprises, ensuring data integrity and consistency. The data is then split into training and testing sets, typically using a ratio of 70%-30% or 80%-20%.

(2) Feature Selection

Feature selection is conducted for both ESG performance data and AI application indicator data to identify relevant feature variables pertaining to the research objective. This process aims to reduce model complexity and computational burden.

(3) Data Preprocessing

Data preprocessing is carried out, encompassing tasks such as data cleaning, handling missing values, feature scaling, and data transformation. These steps ensure data quality and reliability.

(4) Model Training

The SVM model is trained using the designated training set. Throughout the training process, appropriate kernel functions (e.g., linear, polynomial, or radial basis function kernels), regularization parameters, and other hyperparameters are selected. For nonlinear problems, various kernel functions and parameter combinations are explored, and the optimal model is determined through methods such as cross-validation. Assume a training dataset as shown in Eq. (7):

$$\{({x}^{(1)},{y}^{(1)}),({x}^{(2)},{y}^{(2)}),...,({x}^{(m)},{y}^{(m)})\}$$
(7)

In Eq. (7), \({x}^{(i)}\) represents s the input sample, \({y}^{(i)}\) signifies the corresponding label, and \(i=\text{1,2},...,m\).

The kernel function \(\text{K}\left({x}^{(i)}\right),{x}^{(j)}\) is utilized to map input samples to a high-dimensional feature space. Commonly employed kernel functions include the linear kernel function, polynomial kernel function, and radial basis function (RBF) kernel function, among others.

The RBF kernel is identified as particularly effective for addressing the nonlinear characteristics present in the data, leading to its selection for the analysis. Concurrently, the regularization parameter C and the kernel function parameter γ are fine-tuned. The parameter C regulates the penalty imposed on the model for misclassifications, while γ influences the decision boundary created by individual training samples. Through a series of experiments, optimal values for C and γ are determined to achieve a balance between the model’s performance on the training set and its generalization capability on new data, thereby providing a robust predictive model for assessing the ESG performance of central state-owned enterprises.

The objective of SVMs is to determine an optimal hyperplane to separate samples from different classes. This task can be transformed into the following convex optimization problem, as depicted in Eq. (8):

$$\underset{\omega , b}{\text{min}}\frac{1}{2}{\Vert \omega \Vert }^{2}+C\sum_{i=1}^{m}max\left(\text{0,1}-{y}^{\left(i\right)}\left({\omega }^{T}{x}^{\left(i\right)}\right)+b\right)$$
(8)

In Eq. (8), \(\upomega \) stands for the normal vector (hyperplane parameter), b denotes the intercept (bias), C refers to the regularization parameter controlling the complexity of the model, \({x}^{\left(i\right)}\) indicates the input sample, \({y}^{\left(i\right)}\) is the corresponding label, and \(i=\text{1,2},\cdots ,m\).

Optimization algorithms such as gradient descent and coordinate descent can be employed to solve the aforementioned optimization problem, obtaining the optimal normal vector \(\omega \) and b. For a new input sample x, the class to which the sample belongs can be determined by calculating the value of \({\omega }^{T}x+b\), and based on the sign of this value. If the value is greater than 0, it is predicted to belong to the positive class; if it is less than 0, it is predicted to belong to the negative class.

To assess the performance of the model, a test set is commonly utilized, and various evaluation metrics are employed, including accuracy, precision, recall, and F1 score. The formulas for these metrics are as follows:

$$\text{Accuracy}=\frac{TP+TN}{TP+TN+FP+FN}$$
(9)
$$\text{Precision}=\frac{TP}{TP+FP}$$
(10)
$$\text{Recall}=\frac{TP}{TP+FN}$$
(11)
$$\text{F}1\text{ Score}=\frac{2\times \text{Precision}\times \text{Recall}}{\text{Precision}+\text{Recall}}$$
(12)

Here, TP (True Positive) represents the number of true positive instances, i.e., the number of positive class samples correctly predicted as positive; FN(False Negative) represents the number of false negative instances, i.e., the number of positive class samples incorrectly predicted as negative; FP (False Positive) represents the number of false positive instances, i.e., the number of negative class samples incorrectly predicted as positive; TN (True Negative) represents the number of true negative instances, i.e., the number of negative class samples correctly predicted as negative.

Performance evaluation

(1) An analysis of the confusion matrix for the SVM model constructed in this study reveals the following results in Table 4:

Table 4 Confusion matrix results of the SVM model.
Full size table

Subsequently, examination of the numerical outcomes of the evaluation metrics is presented in Fig. 1:

Fig. 1
Fig. 1
Full size image

Numerical results of evaluation metrics for the SVM model.

From the graphical representation, the model’s F1 score stands at 0.878, representing the harmonic mean of precision and recall. The F1 score offers a comprehensive assessment by balancing precision and recall, making it particularly suitable for scenarios involving imbalanced data. In this instance, an F1 score of 0.878 indicates proficient performance of the model in predicting both positive and negative classes. Overall, the model in this example demonstrates robust performance in the prediction task, characterized by high accuracy, precision, recall, and F1 score, thereby showcasing accurate predictions for both positive and negative classes and indicating commendable classification proficiency.

(2) The results of the questionnaire survey are subjected to statistical analysis, with an emphasis on descriptive statistics of the sample and an analysis of group differences. The descriptive statistical findings are illustrated in Fig. 2A–C:

Fig. 2
Fig. 2
Full size image

Descriptive statistical results of the study participants (A) gender, (B) age, (C) education.

From the depicted figures, it is evident that the proportion of males and females is equal, with each gender accounting for 50% of the sample, indicating a balanced gender distribution within the respondents. The majority of respondents fall within the age bracket of 20 to 30 years, comprising 40% of the total sample, reflecting the age composition of the survey population. Individuals aged 30 and above represent 45% of the total sample, indicating a significant proportion of middle-aged respondents. Respondents below the age of 20 constitute 15% of the total sample, possibly comprising students or young professionals who recently entered the workforce. Regarding educational attainment, the majority of respondents hold a bachelor’s degree, constituting 60% of the total sample, suggesting a prevalent presence of undergraduate students among the survey participants. Those with a master’s degree represent 25% of the total sample, indicating a proportion of graduate students. Respondents possessing a doctoral degree or higher account for 5% of the total sample, suggesting the inclusion of individuals from academia or senior professionals.

The distribution of gender, age, and education within the sample exhibits a relatively balanced representation, encompassing respondents from diverse age groups and educational backgrounds. Further statistical analysis of respondents’ backgrounds and roles within the company is presented in Table 5.

Table 5 Respondents’ backgrounds and roles in the company.
Full size table

As illustrated in Table 5, a total of 200 valid questionnaires are collected, representing employees at various levels within central state-owned enterprises, including senior management, middle management, frontline employees, and technical staff. Specifically, 7 senior management personnel participate, accounting for 3.50% of the total sample. This group holds key decision-making and leadership roles within the company and possesses a profound understanding and influence over corporate policies and strategic direction. There are 26 middle management participants, representing 13.00% of the sample. As implementers and coordinators, they exert direct influence over daily operations and employee management. Frontline employees constitute the largest group, with 109 participants, or 54.50%, serving as the foundation of the company’s operations and possessing the most direct experience regarding the organization’s ESG practices. Additionally, 58 technical staff members participate, comprising 29.00% of the sample. They play a crucial role in the company’s technological innovation and application, particularly in the adoption and promotion of AI technologies. This sample distribution provides a comprehensive perspective, allowing for an analysis of AI’s application in ESG practices from the viewpoints of employees at different levels and ensuring the diversity and comprehensiveness of the research findings. Consequently, the sample possesses a certain level of representativeness, enhancing understanding of its characteristics and providing foundational support for subsequent data analysis and research conclusions.

Subsequently, an analysis of questionnaire scores is conducted, assuming a scoring scale of 1–5 for all questions. The scores pertaining to corporate governance (Q1-Q5), environmental protection (Q6–Q10), and social responsibility (Q11–Q15) are depicted in Figs. 3, 4, and 5, respectively:

Fig. 3
Fig. 3
Full size image

Scores of corporate governance dimensions.

Fig. 4
Fig. 4
Full size image

Scores of environmental protection dimension.

Fig. 5
Fig. 5
Full size image

Scores of social responsibility dimensions.

Upon examination of the figures, the average scores for each dimension are as follows: corporate governance: 3.9, environmental protection: 3.7, and social responsibility: 4.2. Correspondingly, the standard deviations are: corporate governance: 0.8, environmental protection: 0.9, and social responsibility: 0.6. The mean score for corporate governance is 3.9 with a standard deviation of 0.8, indicating a generally positive evaluation of corporate governance among respondents, with scores exhibiting relative concentration. For environmental protection, the mean score is 3.7 with a standard deviation of 0.9, suggesting a notable level of concern for environmental issues among respondents, albeit with some divergence in opinions. Regarding social responsibility, the mean score is 4.2 with a standard deviation of 0.6, signifying a prevailing belief among respondents in companies’ effective fulfillment of social responsibilities, with scores demonstrating relative concentration.

Group difference analysis involves comparing the means and standard deviations of key indicators across different groups and conducting statistical tests to derive specific numerical outcomes. To illustrate, the scores of a pivotal question in the questionnaire are examined, with results displayed in Fig. 6:

Fig. 6
Fig. 6
Full size image

Results of group difference analysis.

Upon inspection of the figure, it is apparent that the average score for males is 3.5, with a standard deviation of 1.2, whereas the average score for females is 4.4, with a standard deviation of 1.0. This discrepancy suggests that females exhibit a higher average score on Question 10 compared to males.

Subsequently, a t-test is conducted to scrutinize the disparity in scores on Question 10 between genders, as presented in Table 6:

Table 6 t-Test results.
Full size table

Based on the outcomes of the t-test, the t-value is − 4.37, and the p-value is 0.00002, significantly smaller than 0.05. This indicates a statistically significant difference in scores on Question 10 between males and females. Hence, gender exerts a statistically significant influence on the scores of Question 10.

In the qualitative analysis, content analysis and thematic analysis methods are applied to systematically code and categorize the data, identifying key themes related to the application of AI. The findings of the analysis are summarized in Table 7.

Table 7 Examples of AI application across different ESG dimensions.
Full size table

The study reveals that many central state-owned enterprises have extensively applied AI technologies in the environmental dimension, particularly in resource management and pollution control. Through automated monitoring systems and intelligent optimization technologies, these enterprises have been able to more effectively reduce resource consumption, enhance energy efficiency, and decrease emissions. For example, some enterprises have explicitly reported in their annual reports that AI applications have contributed to reduced energy consumption and the optimization of wastewater treatment, thus improving their environmental performance. In the social dimension, AI applications primarily focus on employee welfare and social responsibility management. By utilizing intelligent employee management systems and data analytics platforms, companies have enhanced employee satisfaction and social responsibility awareness. Notably, some companies have employed AI technologies for health monitoring and safety management, especially in production environments. Predictive maintenance has played a key role in reducing workplace injuries and improving employee welfare. In the governance dimension, AI technologies are widely applied to optimize corporate governance structures, especially in risk management, internal audits, and decision support systems. Through AI-driven predictive models and data analysis tools, companies can more accurately identify potential risks and implement effective early warning and prevention measures. The use of these technologies has significantly improved decision-making transparency and efficiency, which has, in turn, bolstered trust among investors and stakeholders.

(3) Correlation Analysis.

A correlation analysis is performed for all variables, and the results are presented in Table 8.

Table 8 Correlation analysis of variables.
Full size table

Table 8 presents the correlation analysis results, revealing the linear relationships and significance levels between the variables. SDI is positively correlated with both AI and ESG performance, with correlation coefficients of 0.072 and 0.239, respectively, both significant at the 1% level. This suggests that improvements in AI technology adoption and ESG performance contribute to enhanced corporate sustainable development outcomes. Regarding the control variables, company size (Size) is significantly positively correlated with ESG performance (correlation coefficient = 0.201), indicating that larger companies may have more resources and capabilities to achieve stronger performance in environmental, social, and governance areas. The Lev ratio shows a significant negative correlation with SDI, AI, and ESG, implying that higher debt levels may hinder both corporate sustainable development and the adoption of AI technology. Additionally, total AT and RG rate are positively correlated with SDI, with correlation coefficients of 0.203 and 0.281, respectively. This suggests that efficient asset utilization and income growth foster better corporate sustainable development performance. Tobin’s Q is positively correlated with SDI and AI, but negatively correlated with ESG performance (correlation coefficient = -0.081), reflecting a complex relationship between market value and corporate governance performance. These correlations offer valuable insights for the subsequent regression analysis.

(4) Regression Analysis.

The baseline regression results are presented in Table 9. The first column reports the univariate regression results examining the impact of AI usage on SDI. The second column shows the regression results with industry and year fixed effects included, while the third column presents the regression results with control variables incorporated.

Table 9 Baseline regression results.
Full size table

The results from Table 9 indicate that AI usage has a significant positive impact on SDI. In model (1), the regression coefficient for AI usage is 0.002, which is significant at the 1% level, suggesting that AI technology application positively influences sustainable development performance. In model (2), after adding two-way fixed effects for industry and year, the coefficient for AI remains at 0.002, and its significance remains unchanged, confirming the robustness of this result. In model (3), following the inclusion of control variables, the coefficient for AI slightly decreases to 0.001 but continues to be significant at the 1% level, demonstrating that the positive effect of AI on corporate sustainable development performance persists even after accounting for other factors. Among the control variables, Size, total AT, RG rate, and Tobin’s Q exhibit significant positive effects on SDI, while the Lev ratio shows a significant negative effect on SDI. This suggests that corporate resource allocation efficiency, growth capacity, and market performance are key drivers of sustainable development performance. The R2 value of the model increases progressively from 0.005 to 0.370, highlighting that the inclusion of control variables and fixed effects substantially enhances the explanatory power of the model. Overall, the regression results are robust and hold significant academic value.

In summary, listed companies that are larger in size, exhibit lower leverage, higher asset turnover, stronger growth, and better market performance tend to achieve superior sustainable development performance.

(5) ESG Mediating Effect Test.

The regression results for the mediating effect test are presented in Table 10.

Table 10 Mediating effect test results.
Full size table

Table 10 presents the regression results for the mediating effect test, which confirms the mediating role of ESG in the relationship between AI usage and corporate SDI. In Model Path_a, the regression coefficient for AI on ESG is 0.0007, significant at the 1% level, indicating that AI usage significantly enhances corporate ESG performance, thereby satisfying the first condition for establishing a mediating effect. In Model Path_b, the regression coefficient for AI on SDI is 0.204, also significant at the 1% level, suggesting a direct positive effect of AI on corporate sustainable development performance, which meets the second condition. In Model Path_c, which includes both AI and ESG in the regression, the coefficient for AI decreases to 0.0004 but remains significant at the 5% level. ESG shows a significantly positive coefficient, indicating that ESG partially mediates the relationship between AI and SDI. Control variables such as Size, Lev ratio, total AT, and Tobin’s Q also exhibit significant effects on ESG and SDI, further enhancing the explanatory power of the model.

In summary, the regression analysis supports the partial mediating effect of ESG, suggesting that AI not only directly enhances corporate sustainable development performance but also indirectly promotes performance improvement by enhancing ESG performance.

Discussion

This study delves into the impact of AI-driven ESG practices on the sustainable development performance of central state-owned enterprises through questionnaire surveys and factor analysis. The findings reveal generally positive evaluations from respondents towards corporate governance, environmental protection, and social responsibility, with remarkable performance observed in social responsibility.

In terms of corporate governance, respondents perceive central state-owned enterprises positively, with an average score of 3.9 and a standard deviation of 0.8. This aligns with prior research by Napitupulu et al.51, which underscores the role of sound corporate governance in enhancing transparency, decision efficiency, investor confidence, and overall corporate performance. Leveraging AI technologies like data analytics and decision support systems can further refine corporate governance structures, thus improving management efficiency and transparency.

Regarding environmental protection, despite a relatively high average score of 3.7 and a standard deviation of 0.9, there are varied opinions among respondents. This disparity may indicate differences in environmental protection measures and policy execution across different central state-owned enterprises. Ogbeibu et al.52 emphasize that AI applications in environmental monitoring and resource management can significantly enhance a company’s environmental performance. Real-time monitoring and data analysis enable companies to better identify and manage environmental risks, fostering sustainable resource utilization.

In the domain of social responsibility, which garnered the highest score of 4.2 with a standard deviation of 0.6, respondents commend central state-owned enterprises for their efforts and contributions. The clustered scores underscore central state-owned enterprises’ positive performance in social welfare, employee benefits, and community development. According to Lee et al. (2023), companies actively fulfilling social responsibilities are more likely to garner public trust and support, thereby enhancing market competitiveness53. The integration of AI technologies such as intelligent donation platforms and social demand prediction systems can aid companies in better fulfilling social responsibilities, thus enhancing their social image and brand value.

The study’s findings are in line with existing literature, affirming the significant role of AI in enhancing ESG performance. AI technologies facilitate sustainable development by optimizing resource allocation, enhancing management efficiency, and increasing transparency. Specifically, AI’s prowess in data processing and analysis enables companies to more accurately assess and manage environmental and social risks while optimizing governance structures and decision quality.

Despite the social responsibility dimension receiving the highest ratings among respondents, opinions regarding environmental protection displayed greater diversity. This disparity may arise from variations in the implementation of environmental policies and measures within central state-owned enterprises, as well as differing understandings and levels of emphasis placed on environmental sustainability. Challenges faced by central state-owned enterprises in improving their environmental impacts include inconsistencies in technology application, limited financial investment, and the complexities associated with environmental risk management. To address these challenges, central state-owned enterprises must enhance the use of environmental monitoring technologies, optimize resource allocation, and improve the transparency and responsiveness of their environmental management practices. In this context, examples of AI applications can provide valuable insights for improving environmental performance. For instance, AI can facilitate real-time tracking of corporate carbon emissions through advanced monitoring technologies, reduce energy consumption via energy efficiency optimization algorithms, and predict and manage pollution incidents using sophisticated pollution control models. These applications not only assist enterprises in achieving their environmental objectives but also enhance their environmental responsibility and brand image. Through the analysis of specific cases, this study aims to illustrate the practical applications of AI technologies in the environmental domain, offering viable environmental management strategies for central state-owned enterprises.

This study acknowledges that the sample size may limit the generalizability of the research findings, which is a significant consideration. To enhance the representativeness of the results, the respondents span multiple industries, including energy, manufacturing, finance, and technology—sectors that play vital roles within central state-owned enterprises. For example, in the energy sector, AI technologies may focus on optimizing energy consumption and reducing carbon emissions, while in the financial sector, the emphasis may be on risk management and compliance monitoring. This cross-industry sample distribution illuminates the impact of AI on ESG performance in various contexts, providing valuable insights for other central state-owned enterprises. Additionally, when compared to the AI-driven ESG initiatives of non-state-owned enterprises, the advantages of central state-owned enterprises include stronger policy support and resource allocation capabilities. These strengths enable more effective implementation of large-scale AI projects and ESG initiatives. Furthermore, central state-owned enterprises often bear significant national missions and social responsibilities, which provide clearer direction and motivation in their ESG practices. However, notable limitations also exist. Central state-owned enterprises may lack the decision-making flexibility and market responsiveness characteristic of private enterprises, potentially hindering their ability to effectively utilize AI technologies in rapidly changing market environments. Moreover, due to their complex management structures and stringent regulatory requirements, they may encounter bureaucratic obstacles and compliance challenges when implementing innovative technologies. Together, these advantages and limitations create a unique context for central state-owned enterprises within AI-driven ESG initiatives, offering valuable background information for the research findings and highlighting key areas that require attention and optimization in future research and practice.

Building on the proposed avenues for future research, specific recommendations for research methodologies will enhance the understanding of the impact of AI on ESG performance. Conducting longitudinal studies to track changes in ESG performance over time can elucidate the long-term effects of AI. Additionally, qualitative interviews with managers from central state-owned enterprises engaged in AI-driven ESG initiatives can yield in-depth insights into the challenges and successes associated with the adoption of AI technologies. Moreover, investigating the specific impacts of various AI applications, such as machine learning and natural language processing, on ESG performance can provide organizations with concrete insights into the most effective AI technologies for implementing ESG tasks. For instance, in the environmental domain, machine learning algorithms can analyze satellite imagery to monitor deforestation and desertification, enabling companies to better understand and manage their environmental impact. In the social domain, natural language processing technologies can evaluate social media data to assess public perceptions and sentiments regarding corporate social responsibility initiatives, thereby guiding companies in optimizing their social responsibility strategies. In the governance domain, big data analytics can assist organizations in identifying potential compliance risks and internal fraudulent activities, enhancing the transparency and efficiency of corporate governance. These research directions will offer both theoretical and practical guidance for central state-owned enterprises in the implementation of effective ESG strategies.

Conclusion

Research contribution

This study, utilizing survey data, factor analysis, and regression modeling, examines the impact of AI-driven ESG practices on the sustainable development performance of central state-owned enterprises. The findings contribute to the existing literature in several important ways:

First, this study addresses a gap in the literature concerning the relationship between AI and ESG performance, emphasizing the role of AI technology in advancing the sustainable development of central state-owned enterprises. The study demonstrates that AI not only enhances corporate governance and social responsibility by optimizing resource management, improving production efficiency, and fostering innovation, but also promotes comprehensive improvements across ESG dimensions, particularly in environmental management. Although AI technology has yielded notable advancements in corporate governance and social responsibility, particularly in terms of management efficiency and social image, opportunities remain for further improvements in environmental performance. Specifically, areas such as emission reduction and resource optimization continue to present challenges. Future research could explore the full potential of AI technology in these critical areas.

A key contribution of this study is the identification of the mediating role of ESG in leveraging AI to enhance the sustainable development performance of central state-owned enterprises. The regression analysis demonstrates that the adoption of AI technology not only directly improves sustainable development performance but also indirectly facilitates the achievement of sustainability goals by optimizing ESG performance. This finding expands the scope of AI technology’s application in corporate sustainable development strategies and offers a novel perspective for academic research in related fields. Specifically, ESG performance serves as a mediating variable between AI technology and corporate sustainable development, highlighting that, while enterprises pursue economic benefits, they simultaneously enhance their social responsibility, environmental performance, and governance structure, thereby providing a solid foundation for long-term sustainability.

Furthermore, this study contributes new empirical evidence by emphasizing the heterogeneous impact of AI technology on different ESG dimensions. The results indicate that AI’s effect on the environmental dimension remains preliminary and localized, while its application in the social and governance dimensions is more developed and exhibits significant outcomes. By examining the variations in AI application across different enterprises, the study uncovers the complex relationship between AI usage and the enhancement of corporate sustainable development performance, offering valuable insights for both academic and practical understanding of how AI drives corporate sustainability.

This study offers specific practical guidance for policymakers and corporate managers. By thoroughly analyzing the role of AI technology in ESG practices, the study suggests that central state-owned enterprises should not only encourage the adoption of AI technology but also intensify their investment and efforts in environmental protection, particularly in areas such as resource optimization and emissions control. Corporate managers are encouraged to leverage AI technology to improve corporate governance and enhance their social responsibility, thereby achieving a balanced outcome that benefits both economic performance and societal contributions.

In conclusion, this study not only enriches the theoretical understanding of the relationship between AI technology and corporate sustainable development but also provides valuable empirical evidence derived from practical data analysis for central state-owned enterprises in implementing AI-driven ESG strategies. Future research could further investigate the heterogeneous effects of AI technology across various types of enterprises, particularly in different national, industrial, and market contexts, to offer more comprehensive theoretical and practical insights for advancing global sustainable development practices.

Future works and research limitations

This study also acknowledges several challenges and limitations. Firstly, substantial variations exist in the ESG performance among different central state-owned enterprises, which could be attributed to differences in company size, industry characteristics, and geographical factors. Secondly, despite the potential of AI technology, its actual application effects require further validation and optimization. Moreover, the limited sample size of this study, comprising only 200 respondents, may not adequately represent the ESG landscape of all central state-owned enterprises. Future research endeavors should enlarge the sample size and employ a combination of quantitative and qualitative methodologies to delve deeper into the specific impact of AI on ESG performance across diverse contexts. Additionally, careful consideration should be given to the long-term implications and potential risks associated with AI technology in ESG practices to offer more comprehensive guidance and recommendations.